Myoid Gonadal Stromal Tumor

A Clinicopathologic Study of Three Cases of a Distinctive Testicular Tumor

Chia-Sui Kao, MD; Thomas M. Ulbright, MD


Am J Clin Pathol. 2014;142(5):675-682. 

In This Article

Materials and Methods

We reviewed H&E-stained slides from all cases of testicular sex cord–stromal tumors from our surgical pathology files at Indiana University Health Pathology Laboratory between 1990 and 2013 and identified three cases of myoid gonadal stromal tumor that were previously diagnosed as "unclassified sex cord tumor with prominent spindle cells," "unclassified gonadal stromal tumor," and "unclassified spindle cell sex cord–stromal tumor." All were evaluated prior to the study that characterized the immunohistochemical features of myoid gonadal stromal tumor.[4] All three were consultation cases of orchiectomy specimens. The diagnosis was confirmed by the senior author (T.M.U.); we required positive reactivity for both S100 protein and SMA according to the criteria of Du et al,[4] as well as the absence of any sex cord component. All cases showing any sex cord differentiation were excluded since they are more appropriately categorized as unclassified sex cord–stromal tumors. In one case, this was further confirmed by absence of a nested pattern on a reticulin stain.

The tumors were analyzed for various histopathologic features, including size (maximum dimension), necrosis, mitotic rate, cytologic atypia, lymphovascular invasion, encapsulation, and circumscription vs infiltrative margins. Cytologic atypia was defined as a combination of nuclear enlargement, irregularity, pleomorphism, and nucleolar prominence. Additional features assessed were cell shape, nuclear shape, nuclear grooves, perinuclear vacuoles, the quantity and quality of cytoplasm, nature of blood vessels, relationship to the rete testis, and entrapped seminiferous tubules at the periphery.

The results of all immunohistochemical stains performed at the time of diagnosis were recorded, and cases with available material were stained with antibodies directed against S100 protein, SMA, h-caldesmon, desmin, calponin, SOX9, FOXL2, steroidogenic factor 1 (SF-1), inhibin, WT1, and/or calretinin if information was lacking. The antibody sources, dilutions, antigen retrieval methods, and incubations are summarized in Table 1. Immunostains of whole sections of formalin-fixed, paraffin-embedded tissue directed against S100 protein, SMA, h-caldesmon, desmin, calponin, SF-1, inhibin, WT1, and calretinin were conducted using a polymer-based method (EnVision FLEX or FLEX+; Dako, Carpinteria, CA), diaminobenzidine as the chromogen, and a Dako automated immunostaining instrument. Those directed against SOX9 and FOXL2 were conducted using a different polymer-based method (donkey–anti-goat [Jackson ImmunoResearch Laboratories, West Grove, PA] and LSAB2-SA [Dako]), diaminobenzidine as the chromogen, and a Dako automated immunostaining instrument. Negative and positive controls were performed for each immunohistochemical stain. Only nuclear reactivity was considered positive for SOX9, FOXL2, SF-1, and WT1; the presence of cytoplasmic staining without nuclear reactivity was considered negative. Both nuclear and cytoplasmic reactivity was required for S100 protein and calretinin.

All available clinical history and follow-up information were recorded. A review of the previously reported myoid gonadal stromal tumors from the literature (six cases fulfilling our diagnostic criteria) was undertaken.[1–5] This study was approved by the institutional review board at Indiana University.